Effects of fixed orthodontic treatment on nickel and chromium levels in gingival crevicular fluid as a novel systemic biomarker of trace elements: A longitudinal study


Nickel and chromium might induce hypersensitivity. Therefore, they are of interest to orthodontists. Gingival crevicular fluid (GCF) is highly relevant to orthodontic treatments and might reflect systemic changes associated with the inflammatory response induced by orthodontic forces. Therefore, it might also be used to show metal ion changes. Nevertheless, baseline metal levels of GCF are unknown, and the effect of orthodontic treatment on GCF metal levels has not been investigated. The aim of this study was to assess the levels of nickel and chromium in GCF.


Based on a pilot study, the sample size was predetermined as 24 × 3 measurements to obtain test powers above 90%. Nickel and chromium concentrations were measured before treatment and 1 month and 6 months later in 12 female and 12 male patients who had fixed orthodontic appliances using atomic absorption spectrophotometry. The gingival index was also evaluated in each session. The effects of treatment on GCF ions were analyzed using repeated-measures analysis of variance and Friedman tests (α = 0.05, β ≤0.01).


The gingival index worsened over time (chi-square test, P <0.001). The mean nickel levels were 3.894 ± 1.442, 5.913 ± 2.735, and 19.810 ± 8.452 μg per gram, respectively, at baseline, month 1, and month 6. Chromium concentrations were 1.978 ± 0.721, 4.135 ± 1.591, and 13.760 ± 3.555 μg per gram, respectively. Compared with the baseline, nickel increased by 150% and 510%, respectively, in the first and sixth months (Friedman, P <0.0001), and chromium increased by 200% and 700%, respectively (analysis of variance, P <0.0001).


Six months of fixed orthodontic treatment might intensify the levels of nickel and chromium in the GCF as well as gingival inflammation.


  • Nickel and chromium levels in gingival crevicular fluid (GCF) can act as systemic biomarkers.

  • Nickel in GCF might increase approximately 5-fold during the initial stage of orthodontic treatment.

  • Chromium in GCF might increase approximately 7-fold after 6 months of treatment.

  • Fixed orthodontic treatment might exacerbate periodontal conditions.

Orthodontic alloys contain chromium and nickel, which might induce contact allergy, asthma, or hypersensitivity. Corrosion of orthodontic alloys might release nickel and chromium ions into saliva. Chromium oxide forms an anticorrosive passive film over orthodontic appliances. Nevertheless, in clinical situations, this protective layer is disrupted because of mastication, brushing, thermal stresses, saliva flow, biofilm microorganisms and their byproducts and enzymatic activities. recycling of appliances, friction between brackets and wires, occlusal loadings, and acidic drinks, mouthwashes, or toothpastes.

The influence of orthodontic treatment on systemic levels is controversial, showing increases and lack of changes. Systemic exposure can be measured with exposure biomarkers. A medium for systemic exposure with reasonable sensitivity might be gingival crevicular fluid (GCF), a unique biologic exudate that can be collected noninvasively. In an orthodontic setup, GCF might be one of the most relevant biomarkers of exposure because unlike all other biomarkers of exposure (blood, urine, hair), it is directly related to the inflammatory response induced by orthodontic forces. Despite its importance, neither normal metal ion concentrations in GCF nor their levels in orthodontic patients are known.

In view of these shortcomings and controversies, this study was conducted to evaluate the baseline and during-treatment nickel and chromium levels in the GCF of fixed orthodontic patients. As an additional finding, the gingival index was also evaluated.

Material and methods

This longitudinal study was carried out with 72 samples of GCF, taken at 3 times from 24 patients (12 male, 12 female) with fixed orthodontic appliances. The times were (1) pretreatment (baseline), (2) 1 month after the initiation of treatment, and (3) 6 months after baseline.

The research committee of The Azad University approved the protocol and ethics (registered as an orthodontics master’s thesis #2013-t33 for M.S.). Written consents were taken from the participants or their parents. The subjects were sequentially acquired from a list of about 100 patients attending the Department of Orthodontics (in 2011 and 2012), until 2 groups of 12 female and 12 male patients were enrolled.

The inclusion criteria were patients who were willing to be part of the study and needed fixed orthodontic treatment in the maxillary arch. The exclusion criteria were any diseases, syndromes, allergies, metal restorations, consumption of medication or alcohol, smoking, or previous orthodontic treatment. All inclusion criteria had to be fulfilled during the study period.

Only the maxillary arch was treated. Stainless steel brackets (American Orthodontics, Sheboygan, Wis) were bonded using NoMix adhesive (3M ESPE, St Paul, Minn). To reduce the confounders, only nickel-titanium archwires (G&H Wire, Franklin, Ind) were used during the 6-month study (stainless steel archwires were not used). The archwires were replaced with greater sizes about every 4 to 6 weeks, depending on the treatment plan. Within the 6-month period of the study, in 15 patients, all wires were 0.014-, 0.016-, and 0.018-in round wires, used in that order. In 6 patients, the wires were 0.016- and 0.018-in round wires, used sequentially. In 3 patients, the archwires were 0.012-, 0.014-, and 0.016-in round wires followed by 0.019 × 0.025-in square nickel-titanium wires.

Oral hygiene maintenance was taught to the patients before the study, both orally and in written form. Patients were asked to use 1 type and brand of toothpaste (Crest Regular, Procter & Gamble, Cincinatti, Ohio) during the study period. The sampling session was scheduled in the morning. The subjects were told not to brush their teeth on the morning of the sampling session. The patients were instructed, orally and in written form, to avoid consuming nickel- and chromium-rich foods for 48 hours before the sampling visit; they were also asked not to eat or drink on the morning of the scheduled visit.

The sampling was carried out 3 times (before treatment, and 1 month and 6 months later) by a senior orthodontics graduate student (M.S.) trained and calibrated by a periodontist. The teeth and gingivae were not sprayed with water or rinsed to prevent removal of the GCF. To eliminate saliva contamination, the teeth and surrounding gingivae were gently air dried. The region was isolated with cellulose strips. A standardized cellulose acetate absorbent strip with 45-μm micropores (PerioPaper; Oraflow, New York, NY) was used for the GCF collection. It was gently placed in the sulcus for 60 seconds. In each patient, 4 sites for GCF collection were randomly selected in the first sampling session (from the midbuccal sulci and proximal embrasures of the maxillary central and lateral incisors, canines, and premolars) based on the lack of bleeding and the possibility of inserting the PerioPaper strip for at least 1 mm. If a site or tooth showed bleeding on sampling, after the bleeding stopped, other sites or teeth were randomly selected and sampled. In addition, if calculus did not allow inserting the strip, another site was randomly selected. Moreover, if a specific sulcus did not completely wet the strip within 60 seconds (determined macroscopically), the strip was discarded, and a new one was used to sample the gingival fluid from another randomly selected site. After each successful sampling, the PerioPaper strip was removed and placed in a glass container with a lid. Each bottle was filled with 4 strips, each soaked only in 1 site from each patient (no PerioPaper strip was used to sample from more than 1 site). The specimens were kept in a refrigerator at 1°C for a maximum of 1 week and then shipped to the laboratory. The sites of sampling in the first session were recorded for each patient to be sampled in the second and third sessions as well, if they maintained proper conditions for sampling. If the sites were not available, a new sampling site would be randomly selected instead of each site that had bleeding or calculus or insufficient GCF. The strips were weighed before and after GCF sampling (AB204-S; Mettler-Toledo International, Greifensee, Switzerland).

As an additional finding, the gingival index of Loe was used to evaluate and rank gingival health. This index categorizes the health level with the following scores: 0, no inflammation; 1, mild inflammation, slight discoloration, and slight edema, without ulceration or spontaneous bleeding; 2, moderate inflammation, redness, and bleeding on probing; and 3, severe inflammation and considerable hypertrophy, ulceration, and spontaneous bleeding.

The specimens were transferred to the Atomic Energy Organization for atomic absorption spectrophotometry with a calibrated device (AA280Z GTA120; Varian, Mulgrave, Australia). They were then burned at 550°C for 1 hour. The resulting powder was digested by adding hydrochloric acid and placing it in an ultrasonic bath, at the same time. Each specimen was examined 3 times, and the average ion concentration was recorded in parts per million (μg/g of GCF).

Statistical analysis

The sample size was calculated based on a pilot study of 10 patients to obtain test powers greater than 0.9. The post hoc power provided by the final sample size was greater than 99% (α = 0.05, β ≤0.01). The difference between periodontal statuses was assessed using a chi-square test. The correlations between the ion values at both intervals were evaluated with the Spearman correlation coefficient. The D’Agostino-Pearson omnibus normality test confirmed the normality of all groups, except for the nickel levels measured in the first month. Chromium data were analyzed using repeated-measures analysis of variance and the Bonferroni post hoc test with software (version 20.0; IBM, Armonk, NY). Nickel data were analyzed using Friedman and Wilcoxon signed rank tests. The 95% confidence intervals for the means and differences were computed. Gingival index changes over time were evaluated with a chi-square test. The level of significance was set at 0.05 for all tests except the Wilcoxon. It was adjusted to 0.0167 for the Wilcoxon test using the Bonferroni method.


The patients’ mean age was 16.3 ± 3.5 years (range, 13-22 years). There was no significant correlation between the ion levels except for the pair of nickel levels at the baseline and the sixth month ( Fig ; Table I ).

Box plots showing the medians, interquartile ranges, and minimums/maximums (excluding the outliers [ asterisk ]) for nickel ( Ni ) and chromium ( Cr ) levels (μg/g GCF) at baseline ( 0 ), the first month ( 1 ), and the sixth month ( 6 ).

Table I
Results of the Spearman test
Ion Time Statistics Chromium Nickel
Baseline Month 1 Month 6 Baseline Month 1
Chromium Month 1 Rho 0.104
P 0.629
Month 6 Rho 0.173 0.030
P 0.419 0.888
Nickel Baseline Rho −0.057 −0.147 0.072
P 0.793 0.493 0.737
Month 1 Rho −0.030 −0.174 0.364 0.120
P 0.891 0.417 0.081 0.576
Month 6 Rho 0.235 0.202 −0.056 −0.622 0.084
P 0.270 0.344 0.795 0.001 0.695

Compared with the baseline levels, chromium increased about 200% and 700%, respectively, in the first and sixth months of treatment ( Fig ; Table II ). Nickel showed increases of about 150% and 510% compared with the baseline, respectively, in the first and sixth months ( Fig ; Table II ). The repeated-measures analysis of variance and Friedman tests identified the changes as significant (both P values <0.0001). All pairwise comparisons were significant ( Table III ).

Table II
Descriptive statistics for ion levels (μg/g of GCF)
Mean SD CV (%) Min Q1 Med Q3 Max 95% CI
Baseline 1.978 0.721 36 1.010 1.440 1.760 2.513 3.890 1.673 2.282
Month 1 4.135 1.591 38 1.890 2.690 4.125 5.093 8.040 3.464 4.807
Month 6 13.760 3.555 26 7.800 10.500 13.630 16.380 21.000 12.260 15.260
Baseline 3.894 1.442 37 1.100 2.678 3.765 5.025 6.700 3.285 4.503
Month 1 5.913 2.735 46 3.080 4.443 5.105 7.278 16.660 4.758 7.068
Month 6 19.810 8.452 43 8.040 11.250 20.500 26.440 36.090 16.240 23.380
CV , Coefficient of variation; Min , minimum; Q1 , 25th percentile; Med , median; Q3 , 75th percentile; Max , maximum; CI , confidence interval for the mean; Ni , nickel; Cr , chromium.

Table III
Results of Bonferroni test, comparing the estimated marginal means (μg/g of GCF) of the chromium and nickel groups
Ion Time (I) Time (J) Mean diff (I-J) SE P 95% CI diff
Cr Baseline Month 1 −2.158 0.347 0.000007 −3.053 −1.263
Baseline Month 6 −11.782 0.720 0.000000 −13.641 −9.924
Month 1 Month 6 −9.625 0.746 0.000000 −11.551 −7.698
Ni Baseline Month 1 −2.019 0.609 0.001591 −3.590 −0.448
Baseline Month 6 −15.916 1.913 0.000018 −20.856 −10.976
Month 1 Month 6 −13.897 1.884 0.000030 −18.761 −9.032
Only gold members can continue reading. Log In or Register to continue

Apr 4, 2017 | Posted by in Orthodontics | Comments Off on Effects of fixed orthodontic treatment on nickel and chromium levels in gingival crevicular fluid as a novel systemic biomarker of trace elements: A longitudinal study
Premium Wordpress Themes by UFO Themes